The use of 4-dimensional flow magnetic resonance imaging and fluid structure interaction analysis to predict failure of medical therapy in acute uncomplicated type B aortic dissection

4D flow MRI of acute type B aortic dissection.

Central Message.

The novel use of 4D flow MRI and a fluid structure interaction model predicted rapid growth and failure of optimal medical therapy in acute uncomplicated type B aortic dissection.

The optimal management of acute uncomplicated type B aortic dissection (TBAD) is unknown. Historically, optimal medical therapy (OMT) has been the first-line therapy, but endovascular therapy has recently challenged this paradigm due to poor long-term intervention-free survival with OMT.¹ Although high-risk anatomic factors such as total aortic diameter and false lumen (FL) size have been identified,^2,3 novel metrics are necessary to accurately predict OMT failure in uncomplicated TBAD. In this report, we employed 4-dimensional (4D) flow magnetic resonance imaging (MRI) and a fluid-structure interaction (FSI) model to predict early OMT failure in the subacute phase of uncomplicated TBAD (institutional review board No. STUDY00002737; July 23, 2021). Written informed consent was obtained before image acquisition.

Case Report

A 47-year-old man developed acute onset of severe chest and epigastric pain and presented to the emergency room. A computed tomographic angiogram (CTA) demonstrated an acute TBAD with a zone 3 primary intimal tear (PIT) 5.5 cm distal to the left subclavian artery. The maximal descending thoracic aortic diameter was 4.3 cm at the level of the PIT. There were no clinical or radiographic signs of malperfusion or rupture. The patient was treated as an uncomplicated TBAD and managed with OMT. During his index hospitalization, he underwent an additional CTA and a 4D flow MRI (after informed consent), which demonstrated no change in aortic size or morphology. He was subsequently discharged home.

The patient returned for an office visit 3 months after his initial hospitalization, and a repeat CTA demonstrated rapid expansion of his proximal descending thoracic aorta to 6.0 cm. Given the rapid growth of 1.7 cm in 12 weeks, endovascular repair was recommended (Figure 1). The patient underwent thoracic endovascular aortic repair (TEVAR), had an uneventful postoperative course, and was discharged home on postoperative day 4. A postoperative CTA 1-month following TEVAR demonstrated a stable total aortic diameter of 5.5 cm, with complete false lumen thrombosis (Figure E1).

Figure 1.

Computed tomography angiography (CTA) of rapidly expanding type B aortic dissection (TBAD). A, Initial CTA axial and sagittal multiplanar reconstruction (MPR) demonstrating TBAD with a maximum aortic diameter of 4.3 cm. B, Follow-up CTA axial and sagittal MPR at 3 months demonstrating a TBAD with interval growth of descending aorta measuring 6.0 cm. C, Three-dimensional cinematic rendering of TBAD-associated aneurysmal degeneration at 3 months.

Figure E1.

Computed tomography angiography (CTA) 1-month after thoracic endovascular aortic repair (TEVAR). A, 3D volume rendering of the thoracic aorta post-TEVAR. B, Axial CTA demonstrating complete false lumen thrombosis. C, Sagittal CTA demonstrating complete false lumen thrombosis.

4D Flow MRI and FSI Modeling

On the 4D flow MRI, the peak velocity through the PIT was 1.80 m/second, which is the highest velocity (mean, 1.2 ± 0.5 m/second) in our institutional 4D flow MRI database of acute uncomplicated TBAD (Figure 2, A). In addition, there was substantial retrograde flow in the FL proximal to the PIT with 2 distinct vortical structures above and below the tear. The FL wall shear stress (WSS) directly across from the PIT was in the top 5% of total estimated WSS (>1 Pa) observed in the aorta (Figure 2, A).

Figure 2.

Four-dimensional flow magnetic resonance imaging of acute uncomplicated type B aortic dissection during index hospitalization and aortic wall stress and growth rate. A, Streamlines of blood flow through the true and false lumens. The peak velocity through the primary intimal tear is 1.8 m/second. B, Wall shear stress (WSS) of the true and false lumen demonstrating high WSS on the false lumen wall opposite the primary intimal tear. C, Structural aortic wall stress from fluid structure interaction (FSI) model demonstrating high stress on the false lumen wall opposite the primary intimal tear at the initial computed tomography angiography (CTA) scan. D, Growth rate heat map generated from baseline and 3-month follow-up CTA demonstrating highest growth rate in the false lumen wall opposite the primary intimal tear. MRI, Magnetic resonance imaging; PIT, primary intimal tear.

Three-dimensional geometries of the dissected thoracoabdominal aorta were segmented from the baseline CTA scan. Individual computational meshes were generated for the aortic wall and the blood to perform 1-way FSI simulations (Figure E2). The FSI analyses were performed to estimate aortic wall stress distribution in the following 2 steps: computational fluid dynamics simulation was performed and flow rates at the inlet and outlets were extracted from the 4D flow MRI data (Figure E2) and the nonuniform pressure distribution of the true and false lumens extracted from computational fluid dynamics was applied to the aortic wall to calculate the aortic wall stress distributions using finite element analysis (Figure 2, C). The simulated peak velocity of 1.88 m/second at the PIT matches well with the 4D flow MRI peak velocity of 1.80 m/second through the PIT. Additionally, distributions of aortic growth rate (Figure 2, D) were quantified using the initial and the 3-month follow-up CTA scans. To quantify their underlying spatial correlation, the aortic wall stress contour, and a growth rate heat map were registered to generate scatter data plots. Linear regression analysis revealed a positive slope (1.22% × 10⁻⁴ %/year × Pa; P = 3.41 × 10⁻⁶) between wall stress and growth rate distributions. The positive slope confirmed that the FL location with the highest wall stress correlated with the location of the highest growth rate (Figure E3).

Figure E2.

Fluid structure interaction model. A, Three-dimensional (3D) computational mesh of the aortic wall. B, 3D computational mesh of the blood flow. C, Bar graph and table showing equivalent flow rates extracted from the 4-dimensional (4D) flow magnetic resonance imaging (MRI) and used for the fluid structure interaction (FSI) model. BCA, Brachiocephalic artery; LCCA, left common carotid artery; LSA, left subclavian artery; SMA, superior mesenteric artery.

Figure E3.

Linear regression analysis demonstrating a high correlation between aortic wall stress and growth rate, highest in the false lumen near the left subclavian artery. FSI, Fluid structure interaction.

Discussion

Data continue to emerge regarding the poor long-term outcomes of uncomplicated TBAD treated with OMT. Our institutional data from 146 patients with uncomplicated TBAD demonstrated an intervention-free survival rate of 50% at 5 years and 32% at 10 years.¹ It is widely recognized that TEVAR is more effective in remodeling the aorta in the acute rather than chronic phase of TBAD, due to the compliance and elasticity of the dissection flap in the acute phase. However valid arguments exist against treating all TBADs with TEVAR, including cost, periprocedural risks of stroke, death, paraplegia, and vascular injury as well long-term sequelae of TEVAR-induced aortic stiffening, including adverse cardiac remodeling and the development of hypertension.⁴ Ideally, patients at high risk of OMT failure could be identified in the acute phase and treated with early TEVAR.

In this report, we conducted a novel patient-specific investigation of the FL in acute TBAD using 4D flow MRI and FSI analysis. The 4D flow MRI data demonstrated a high velocity of flow through the PIT resulting in high FL WSS opposite the PIT. The blood flow data informed the FSI model, which was combined with 3-dimensonal aortic geometries to generate heat maps of FL wall stress and growth rates. These methods demonstrated a positive correlation (P < .0001) between aortic wall stress and growth rate.

Although other groups have published 4D flow MRI results in the setting of acute TBAD,⁵ this case represents the first time that the combined approach of 4D flow MRI and FSI has been used to predict aortic growth in acute uncomplicated TBAD. Although numerous anatomic features (eg, aortic diameter, PIT size, and FL patency) have been identified as potential predictors of OMT failure in TBAD,^2,3 these parameters do not provide critical hemodynamic and biomechanical information about the forces acting upon the FL wall. In this case, the velocity of the blood flow jet through the PIT and the determination of structural stress on the FL wall correctly predicted rapid FL growth of 1.7 cm in 3 months, resulting in a change in therapy from OMT to TEVAR. The combination of sophisticated blood flow imaging and computational modeling of the structural stress on the FL wall offers a novel paradigm to identify TBAD patients at high risk for OMT failure in the acute phase and triage these patients for early endovascular therapy, which may improve overall long-term survival.

Conflict of Interest Statement

Dr Leshnower is a speaker for Medtronic and a consultant for Endospan Inc. All other authors reported no conflicts of interest.

The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.